PROJECT SUMMARY Improving diagnosis of tuberculosis (TB) is critical to reducing its burden among children. Key challenges include that many children do not produce sputum and, even among those who do, bacillary burden in sputum is often low. Therefore, identification of a non-sputum based test with high accuracy for diagnosing TB in children is a key global health priority. One approach is to evaluate volatile organic compounds (VOCs) in exhaled breath, which is easily collected from children of all ages. Recent research has identified at least two VOCs ? methyl nicotinate and methyl p-anisate ? that are a byproduct of Mycobacterium tuberculosis metabolism and should therefore be present in anyone with active TB, regardless of age. These VOCs do not appear to be emitted by other bacteria of the respiratory tract; do not appear to be found at high concentrations in the ambient environment; and have been identified at high concentrations in the breath of TB patients but not healthy controls. However, the lack of a technology to easily measure specific VOCs at the point-of-care has been a key barrier to further validation and use of these VOCs as a biomarker for TB diagnosis. The overall objective of this proposal is to validate methyl nicotinate and methyl p-anisate in exhaled breath as biomarkers for diagnosis of intra-thoracic TB in children when detected using a novel, low-cost, handheld device. The device is a solid-state sensor based on metal-functionalized 3D titanium dioxide (TiO2) nanotube arrays that bind specific VOCs when a specific voltage gradient is applied. Binding results in a change in current that is proportional to the concentration of target VOCs present in the breath. The sensor has been customized to specifically bind methyl-nicotinate and methyl p-anisate and has a low limit-of-detection. Furthermore, the detection is fast (on the order of minutes), reagent-free, and requires no sample preparation. Our central hypothesis is that the breath sensor will increase the proportion of confirmed TB diagnoses in children compared to the currently recommended molecular assay (Xpert MTB/RIF, Cepheid, USA). In Aim 1, we will further develop the novel breath sensor into a field-ready childhood TB diagnostic platform by 1) incorporating positive and negative controls into the existing sensor; 2) designing a prototype for direct collection and delivery of a breath sample to the sensor (versus collection of breath in a breath bag and connecting the bag to the sensor, as is currently done); and 3) performing usability testing to refine the device prototypes. In Aim 2, we will validate the accuracy of the two candidate biomarkers and breath sensor device for diagnosis of intra-thoracic TB in children. We will enroll and follow 700-875 children with a clinical suspicion of TB and 25 children in each of four control groups. We will assess diagnostic accuracy of the breath sensor in reference to NIH case definitions for pediatric TB and incorporate data from children in the control groups to verify specificity and evaluate alternate cut-points. Completion of these aims will result in high-quality data on the performance of a novel breath technology for non-invasive diagnosis of TB in children.